Alterations in Religious Rituals Due to COVID-19 Could Be Related to Intragroup Negativity: A Case of Changes in Receiving Holy Communion in the Roman Catholic Community in Poland

Abstract

The COVID-19 pandemic has affected various domains of everyday life, including important religious rituals. In the Roman Catholic Church in Poland, the reception of Holy Communion was substantially altered. The suggestion of the Polish Episcopal Conference and diocesan bishops was to receive Holy Communion on the hand during the pandemic, while receiving on the tongue had been the default form before the pandemic. The present studies investigated whether alterations in the form of receiving Holy Communion during the pandemic resulted in intragroup negativity. A total of 376 Polish Roman Catholics participated in two online studies. The most ambivalent emotions toward their religious community were experienced by the followers who recognized reception of Holy Communion on the hand only. Intergroup bias occurred within the “hand only” and the “mouth only” groups and consisted in out-group favoritism (within the “hand only”) and out-group derogation (“mouth only”) in their perception of religious orientation. Intergroup empathy bias occurred in the “hand only” and “spiritual reception” groups, which reported less empathy toward those of the out-group (“mouth only”) infected with SARS-CoV-2. The highest legitimacy of the Church authority was agreed upon by the supporters of both forms of receiving Holy Communion.

1. Introduction

The COVID-19 pandemic, declared so by the WHO on March 11th, has significantly affected large populations in terms of serious social, political, economic, and psychological aspects (Holmes et al. 2020). Disruption to people’s routines and the inability to freely navigate daily life have resulted in serious consequences for wellbeing, mental health, and social relationships (Duan and Zhu 2020; Zhang et al. 2020). The pandemic has also had a significant impact on the religious aspect of social life (Sulkowski and Ignatowski 2020).

The pandemic caused unprecedented alterations in religious rituals and services (e.g., suspended public worship services during the Holy Week in the Roman Catholic Church; Vatican News 2020), which resulted in a sense of fear and isolation among the faithful (Parish 2020). The suggested form of receiving Holy Communion was one of the most important examples of ritual alterations due to the COVID-19 mitigation guidelines within the Roman Catholic Church in Poland (Sulkowski and Ignatowski 2020). The Church authorities strongly suggested reception of Holy Communion on the hand in order to prevent the spread of the virus, whereas the default form before the pandemic (i.e., in the mouth) was allowed only in particular cases (Polish Episcopal Conference 2020a). Although both forms of receiving Holy Communion are approved by the Catholic Church (General Instruction of the Roman Missal 2021; No. 160.), in Poland, the administration of consecrated host was advised to be on the tongue (Polish Episcopal Conference 2005).

The altered instructions regarding receiving Holy Communion resulted in conflicts among Roman Catholics in Poland. Billboards including exhortations to stop receiving Holy Communion on the hand appeared in Polish streets (e.g., suggesting that the hands of people who receive Holy Communion to the hand are dirty; deon.pl 2020). Sermons indicating that receiving Holy Communion on the hands is blasphemous started to be shared via social media. In order to respond to the criticism regarding receiving Holy Communion on the hand, the Polish Episcopate took another official stand on this issue in October 2020 (Polish Episcopal Conference 2020b). However, the manifestations of criticism toward the altered form of receiving Holy Communion during the pandemic may indicate that this modification of ritual may have become a condition that led to intragroup conflict (Jehn 1997) among Roman Catholics in Poland.

This possible conflict is particularly worth studying in Poland due to the cultural context. Polish identity is perceived to be strongly associated with Catholicism (Bobrowicz and Nowak 2021; Millard 1997). However, religious identity is highly fragmented at the individual level of believers of the Catholic Church in Poland (Topidi 2019), and at the social level (Zubrzycki 2020). These divisions, in a very simplified sense, can be defined as a conflict between “progressive” and “traditional” Catholics. This diversity may strongly affect reactions to alterations to rituals. Particularly, traditional Catholics can resist attempts to alter rituals and can treat them as an attack on their most basic religious values. Progressive Catholics, by contrast, can criticize traditionally oriented Catholics who present such attitudes due to their blind faith in the outer form of ritual without understanding its internal meaning.

The goals of the present studies were to investigate how beliefs about adherence to alterations to the ritual of receiving Holy Communion affected relationships within the religious community during the pandemic. In the first study, we investigated how membership to a group of a particular belief connected with receiving Holy Communion during the pandemic (“mouth only” vs. “hand only” vs. “both forms accepted”) was related to experiencing positive and negative affect toward the whole religious community, and the legitimacy of the Church authority, and whether it predicted intergroup bias (i.e., favoritism toward the in-group or derogation toward the out-group; Hewstone et al. 2002). In the second study, we investigated whether membership to a group of a particular belief connected with receiving Holy Communion during the pandemic was related to intergroup empathy bias (Cikara et al. 2014), namely, dampened empathy toward persons sharing different beliefs connected with receiving Holy Communion.

1.1. Individual and Interpersonal Dimensions of Religiosity

Alterations to religious rituals and services due to the COVID-19 pandemic are an important challenge for the individual religious experience, but also for social relationships among religious followers and the church hierarchy. Thus, individual dimensions of religiosity can affect decisions regarding alterations to rituals and the perceptions of other followers’ decisions in this case.

Religious orientation theory (Allport and Ross 1967; Gorsuch 1988) distinguishes between intrinsic orientation toward religion (i.e., religion is deeply personal to the individual) and extrinsic orientation toward religion (i.e., the person puts emphasis on protection, consolation, social status, and group participation). Extrinsic religious orientation refers to personal (religion as a source of comfort) and social (religion as a source of social participation and social gain) aspects of religiosity (Maltby 2002). The authority of the clergy is another important dimension of religiosity (Falbo et al. 1987). Various versions of the legitimacy theory predict that the duty and obligation to obey legitimate authorities generally trump people’s personal moral and religious values (Skitka et al. 2009).

The alteration in the reception of Holy Communion could have resulted in spiritual doubts among individuals with high intrinsic religious orientation (e.g., whether alteration in receiving Holy Communion due to fear of COVID-19 transmission can be a sign of a weak faith). Previous studies demonstrated that religious people did not necessarily adhere to COVID-19 mitigation guidelines. During the introduction of the COVID-19 mitigation policy in the United States, religiosity in a community led to decreased adherence to shelter-in-place directives (DeFranza et al. 2020). High individual religiosity, therefore, could prompt one to act against rules (e.g., sanitary rules) that interfere with or violate religious values.

Extrinsic religious orientation and the authority of the clergy could have also been challenged due to the alteration in religious rituals during the pandemic. Previous studies demonstrating the psychological sense of community and identification with one’s religious group were correlated with personal well-being (Obst and Tham 2009). However, concerns about negative experiences with religious people or institutions and interpersonal conflicts related to religious issues may imply a sense of religious and spiritual struggle (Exline et al. 2014). Religious individuals may experience stress when they face hypocrisy from believers and the clergy (Krause et al. 2000). Disagreements among group members about interpersonal issues, such as differences in norms and values, could also result in intragroup relationship conflicts (Jehn 1997). Recent meta-analyses have demonstrated that relationship conflicts among group members are related to lower trust, cohesion, satisfaction, commitment, and identification with in-groups (De Dreu and Weingart 2003; De Wit et al. 2012).

The differences in reactions to the altered ritual during the pandemic caused adverse interpersonal results (e.g., appearance of billboards and sermons suggesting that receiving Holy Communion on the hand is sinful). These consequences may consist in interpersonal doubts (e.g., whether to adopt the alteration in receiving Holy Communion in order to obey the church hierarchy or in order to protect others) or adverse interpersonal perception (e.g., the hostile attributions of others’ behaviors that are incompatible with one’s own attitude). This could create a serious threat for both the sense of religious community and the legitimacy of the religious hierarchy.

1.2. Alteration of Rituals Due to the COVID-19 and Intragroup Conflict in the Religious Community

The episcopal and diocesan instructions on the proper reception of Holy Communion in the pandemic resulted in a division of Catholic followers into three groups, i.e., those who recognized reception (a) in the mouth only (b) on the hand only and (c) those who accepted both forms of receiving Holy Communion during the pandemic or who were practicing “spiritual communion” (receiving Holy Communion spiritually without any external behavior). As a result, the followers of the same religious community could be treated as sharing the same beliefs (in-groups) or different beliefs (out-groups), which could result in intergroup processes.

Previous research demonstrated that religious individuals tend to help those who share their values (Preston et al. 2010; Saroglou 2006) and tend to less favorably perceive members of nonreligious out-groups and those who act against their values (Hunter 2001; Jackson and Hunsberger 1999; Johnson et al. 2012; LaBouff et al. 2012; Verkuyten 2007). The pattern of positive attitudes toward similar religious others (i.e., in-group favoritism) and negative attitudes toward nonreligious others (i.e., out-group derogation) is referred to as religious intergroup bias (Johnson et al. 2012).

Religious intergroup bias is a particular example of general intergroup bias, which refers to the tendency of individuals to evaluate the in-group (any group they belong to) and its members more positively than the out-group and its members (Hewstone et al. 2002; Tajfel and Turner 1986). The bias was stronger when the group identification was more pronounced (Demoulin et al. 2009; Mirosławska and Kofta 2007). The various forms of intragroup bias included e.g., infrahumanization (subtle dehumanization of out-groups by treating them as less human; Leyens 2009) and the intergroup empathy bias (the tendency not only to empathize less with out-group relative to in-group members, but also to feel pleasure in response to their pain and pain in response to their pleasure; Cikara et al. 2014). The consequences of such intergroup biases are severe (e.g., diminished prosociality, increased antisociality, moral outrage toward out-groups; cf. Haslam and Loughnan 2014).

Previous studies demonstrated that intergroup bias (favoritism and derogation) may indeed exist within the same religious group. Swan et al. (2014) demonstrated that the degree of belief and indicators of dogmatic thinking moderated the effect of favorable perception of a member of the religious in-group. That study showed that more positive evaluations of a target religious person depended on the assumption of the target person’s strong but flexible belief.

The attitude toward religious in-groups who did not share the same beliefs about the proper form of receiving Holy Communion in the pandemic can be explained from the point of view of costly signaling theory (Sosis 2005). Religion often involves costly signals such as elaborate rituals, and such behaviors are understood to communicate commitment to one’s in-group (Sosis and Alcorta 2003). Giving to religious charities or adhering to religious dietary restrictions, as costly signals of religious commitment, were related to increased trust within the religious group (Hall et al. 2015). Receiving Holy Communion in the mouth might be understood as a costly signal of commitment to the religious group, while changing the form of reception of Holy Communion during the pandemic might be perceived as a signal of decreased commitment to the religious group.

Predictions about the intragroup consequences of ritual alteration due to COVID-19 could be also drawn from the theory of ritual alteration as moral violation (Stein et al. 2021b). Stein and colleagues demonstrated that because group rituals symbolize sacred group values, even minor alterations to them provoke moral outrage and a tendency to punish the violators. Individuals who were more committed to their religion (i.e., Judaism) and strongly believed that the ritual (Passover) symbolizes religious values experienced the most outrage and willingness to punish the violation (Stein et al. 2021b; Study 5). Alterations of a ritual represent a particular moral violation to ingroup members, not outgroup members, because outgroup members do not share the group’s values (Stein et al. 2021a). Hence, differences in adherence to altered ritual within the religious community may result in the division of the community into ingroups (sharing values underlying the traditional form of ritual) and outgroups (people who adhere to altered ritual). This division may lead to mutual negativity within members of the same religious community.

The present study focused on the social psychological processes. However, the alteration of receiving Holy Communion during the pandemic can be also analyzed in the context of ritual studies. Rituals are defined as “predefined sequences of action characterized by rigidity, formality, and repetition that are embedded in a larger system of symbolism and meaning” (Stein et al. 2021b, p. 4; see also Hobson et al. 2018). Ritual often serves as a means of establishing group identity (Hüsken 2007; Kakar 2010). Social groups rely on ritualized activities for social cohesion and control (Bell 2009). Thus, alteration of ritual can challenge these social functions of ritual. DuBois et al. (2010) indicated that religious groups tend be highly critical of those who are not involved in a ritual or who drop out of the ritual. Explicit critique and accusing others of having made mistakes in ritualistic activity is a common practice (Hüsken 2007).

Although rituals are perceived as rather static and unchanging, most of them do in fact undergo changes (e.g., in the course of time, as a result of their transfer to another cultural context, or because they are ‘updated’ to meet the requirements of changed circumstances; Hüsken 2007). There are various types of deviation from an original or earlier version of a ritual (Grimes 1990) and some of these deviations can be judged negatively (e.g., as ritual failures). Ritual failures refer to cases in which a ritual is performed imperfectly, giving rise to its challenging and negotiation in relation to the ritual community’s developing identity (Hüsken and Neubert 2012). The alteration of rituals due to the preventive measures imposed during the pandemic can trigger a process of ritual change or negotiation. While some members of the religious group can judge the alteration of receiving Holy Communion as justified by the circumstances and as not altering the meaning of the ritual, another part of the religious community may perceive this alteration as ritual violation or misexecution (Grimes 1990). This division can cause intergroup negativity within the religious community.

1.3. The Present Study

The goals of the first study were to investigate how the individual’s beliefs about the proper form of reception of Holy Communion affected (a) affective reactions toward the religious group as a whole (Roman Catholics); (b) the legitimacy of the Church authority in Poland; and (c) perceptions of the members of the religious group who did not share the individual’s beliefs about the proper form of reception of Holy Communion during the COVID-19 pandemic. The choice to obey sanitary suggestions connected with receiving Holy Communion on the hand may have resulted from the fear of infection (Clark et al. 2020) and may have been also related to more anger, less trust, and commitment toward those followers who did not obey the instructions. Thus, we hypothesized that the participants who held the beliefs that Holy Communion should be received on the hand would be the most affectively ambivalent (i.e., they would report the highest level of negative emotions, such as fear and anger, and the lowest level of positive emotions, e.g., trust) toward their religious group compared to the individuals who recognized reception in the mouth or both forms as proper. Second, Polish episcopal and diocesan church authorities suggested that Holy Communion should be received on the hand. As a result, it was hypothesized that legitimization of the Church authority would be higher among the followers who considered the reception Holy Communion on the hand as the correct form or among those who accepted both forms. Thirdly, we hypothesized that in the “hand only” and “mouth only” groups, intragroup conflict would arise, and intergroup bias would occur. We investigated in-group favoritism or out-group derogation in the ascription of religiosity (measured as intrinsic and extrinsic orientation to religion). We predicted that members of the in-groups would be perceived as more religious (they would obtain a more favorable assessment on both intrinsic and extrinsic orientation to religion) compared to the members of the out-groups. Alternatively, according to costly signaling (Sosis 2005), it could be predicted that the “mouth only” group would be perceived as more religious compared to the “hand only” one due to maintaining the costly signal (receiving Holy Communion in the mouth) despite the sanitary restrictions. Since for the individuals accepting both forms of reception of Holy Communion, other participants were always treated as in-groups, we did not predict intergroup bias in this particular group. However, the “mouth only” group should be perceived as more religious if reception of Holy Communion is treated as a costly signal.

The goals of the second study were to investigate whether differences in the individual’s beliefs about the proper form of reception of Holy Communion may result in intergroup empathy bias. In the second study, we included another subgroup according to the form of receiving Holy Communion: Individuals who maintained spiritual reception of Holy Communion during the pandemic. In order to examine empathic responding toward in-group and out-group, we used vignettes that described three target persons infected with SARS-COV-2, including a manipulation of the target person’s belief about the proper form of reception of Holy Communion (“on the hand only” vs. “in the mouth” vs. control condition). Again, we hypothesized that in the “hand only” and “mouth only” groups, the intergroup empathy bias would occur. We predicted that members of in-groups would be more compassionate toward a target person who had received Holy Communion in the way that was similar to their belief compared to the members of the out-groups. For the persons accepting both forms of the reception of Holy Communion, both target persons described in vignettes as receiving Holy Communion on the hand instead of in the mouth were always treated as in-groups, thus we did not predict the intergroup bias in this particular group. Similarly, for the persons preferring the spiritual form of the reception of Holy Communion, both target persons described in vignettes as receiving Holy Communion on the hand instead of in the mouth were always treated as out-groups, thus we did not predict intergroup bias in this particular group.

2. Study 1

2.1. Materials and Methods

2.1.1. Participants and Procedure

One hundred and ninety-seven individuals (163 women and 34 men) strongly involved in religious practices (teaching theology or joining religious communities) participated in the present study. The age of the participants ranged from 17 to 65 years (M = 32.9; SD = 10.5). All of them attended religious practices at least once a week. One hundred and fifty participants reported higher education (76.1%), 46 reported secondary education (23.4%), and 1 person reported primary education. All the participants were Roman Catholics. The sample size was determined a priori using G*Power (Faul et al. 2007). The required sample size to detect a small effect size (f² = 0.10) in ANOVA with repeated measurements and interactions of between and within factors (power = 0.80; α = 0.05) was computed as N = 174. Thus, the number of participants in the present study met this criterion. The study was approved by the institutional ethics board (KEUS.98/02.2021).

2.1.2. Measures

The following measures were used:

• Intrinsic and extrinsic religious orientation: The age universal I-E scale-12 (Maltby 1999) was used to assess intrinsic and extrinsic religious orientation. The scale consists of 6 items measuring intrinsic religious orientation (e.g., “I try hard to live all my life according to my religious beliefs”) and 6 items measuring extrinsic religious orientation (e.g., “What religion offers me most is comfort in times of trouble and sorrow”). The reliability of the intrinsic religious orientation was α = 0.741, and the reliability of the extrinsic religious orientation was α = 0.811.
• Positive and negative emotions toward the religious community: We used five indicators of positive emotions toward the religious community (trust, care, closeness, commitment, and joy) and four indicators of negative emotions toward the religious community (anger, contempt, fear, sadness). The participants rated how frequently they felt such emotions toward their Church members on a Likert-type scale ranging from 0 (Never) to 4 (Very often). The principal factor analysis yielded a clear two-factor solution (explained variance = 61.455%), with the factors of positive emotions (loadings > 0.702) and negative emotions (loadings > 0.657). The reliability of positive emotions toward the religious community was α = 0.838, while the reliability of negative emotions toward the religious community was α = 0.747.
• Perceived legitimacy of Church authority: It was measured with three items adopted from Van der Toorn et al. (2015). The items were reworded and were as follows: “I feel I should accept the decisions made by my Church authorities, even when I think they are wrong”; “I think that it hurts my religious group when I disagree with my Church authorities”; and “I feel that it is wrong to ignore the instructions of my Church authorities even when I can get away with it”. The responses were made on a 5-point scale ranging from 0 (Disagree strongly) to 4 (Agree strongly). The reliability of the measure was α = 0.823.
• In-group and out-group perceived religious orientation: The study participants were asked to rate how statements regarding intrinsic and extrinsic orientation to religion were characteristic of persons who received Holy Communion on the hand or in the mouth. These sentences were adopted from the Age Universal I-E scale-12 (Maltby 1999) and reworded in order to assess the perception of a particular group. The participants used a Likert-type scale ranging from 0 (Very bad) to 4 (Very good), with the mid-point of 2 (Hard to tell). The reliability of each scale (intrinsic vs. extrinsic) for each target (on the hand vs. in the mouth) ranged from α = 0.857 (extrinsic orientation ascribed to the “hand only” group) to α = 0.947 (intrinsic orientation ascribed to the “hand only” group).

2.2. Procedure

The participants were invited to the study by means of information posted in social media groups gathering people involved in religious movements and teaching religion. They were also invited to share the invitation with other persons. After clicking the link to the study, the participants were informed about the aim of the questionnaire and were asked to accept the terms of participation. First, the participants assessed their religious orientation. Next, they were asked to indicate which form of the reception of Holy Communion was proper (“hand only” vs. “mouth only” vs. “both forms accepted”) and to explain their approach (in order to enhance their identification with the beliefs). In the next sections, the participants were asked to assess intrinsic and extrinsic religious orientation of persons who received Holy Communion on the hand and in the mouth. These two assessments were divided by a filler assessment of polychronicity. Afterwards, the participants were asked about the emotional reactions toward the members of their Church, the legitimacy of the Church authority, and the demographic data. On completing the study, the participants were debriefed by means of announcements posted in the same social media in which the invitations had been posted. The authors of the study also answered all the questions which had emerged during the study.

2.3. Results

2.3.1. Descriptive Statistics

Means, standard deviations, and correlation among the study variables are given in

Table 1. Descriptive statistics for the study 1 variables.

Variable	1	2	3	4	5	6	7	8	9
1. Intrinsic RO
2. Extrinsic RO	0.328 ***
3. PA	0.315 ***	0.414 ***
4. NA	−0.031	0.092	−0.125
5. LCA	0.306 ***	0.219 *	0.271 ***	−0.134
6. ”hand only” (IRO)	0.259 ***	0.247 ***	0.191 **	0.053	0.245 ***
7. “hand only” (ERO)	0.211 **	0.532 ***	0.271 ***	0.024	0.150 *	0.507 ***
8. “mouth only” (IRO)	0.266 ***	0.402 ***	0.356 ***	−.040	0.160 *	0.669 ***	0.466 ***
9. “mouth only” (ERO)	0.139	0.469 ***	0.188 ***	0.091	0.115	0.509 ***	0.576 ***	0.654 ***
M	3.376	2.183	2.301	1.641	2.426	2.662	2.276	2.809	2.443
SD	0.539	0.884	0.876	0.982	1.153	0.845	0.697	0.832	0.784
Age	0.234 **	−0.051	0.289 ***	−0.002	0.049	0.089	−0.034	0.108	0.026
Gender	0.145 *	0.181 *	0.016	0.045	0.095	0.247 ***	0.111	0.076	0.147 *

Note. Gender was coded: 0—men; 1—women. RO—religious orientation; PA—positive affect; NA—negative affect; LCA—legitimacy of the Church authority; IRO—intrinsic religious orientation; ERO—extrinsic religious orientation. * p < 0.05; ** p < 0.01; *** p < 0.001.

.

Intrinsic and extrinsic religious orientations were positively correlated with each other, and positively related to positive affect toward the Catholic Church members, the legitimacy of the Church authoritym and ascriptions of intrinsic and extrinsic religious orientations to the “hand only” and “mouth only” groups. Age and gender were also correlated with religious orientation and with positive affect toward Church members. Thus, in the next analysis, individual intrinsic and extrinsic religious orientation, age, and gender were controlled for.

In terms of the question related to the proper form of reception of Holy Communion, three groups were distinguished: “hand only” (N = 72; 36.55%), “mouth only” (N = 49; 24.87%), and “both forms accepted” (N = 76; 38.58%). We conducted 3 between-group (“hand only” vs. “mouth only” vs. “both forms accepted”) and 2 within-group (intrinsic vs. extrinsic) ANCOVAs on religious orientation. Women reported higher religious orientation compared to men (F(1, 192) = 243.421; p < 0.001; η_p² = 0.045). Intrinsic religious orientation was reported as higher compared to extrinsic religious orientation (F(1, 192) = 12.553; p < 0.001; η_p² = 0.061). Religious orientation was also predicted by group membership (F(2, 192) = 4.387; p = 0.014; η_p² = 0.044). The “mouth only” (M = 2.871; SD = 1.070) and “both forms accepted” groups (M = 2.868; SD = 0.878) showed higher religiosity compared to the “hand only” group (M = 2.624; SD = 0.896; p < 0.080). The interaction between group membership and the type of religious orientation (intrinsic vs. extrinsic) was non-significant. Thus, in the subsequent analyses, individual intrinsic and extrinsic religious orientations were controlled for.

2.3.2. Group Membership and Emotional Reactions toward the Religious Community

First, we investigated whether group membership (“hand only” vs. “mouth only” vs. “both forms accepted”) determined the (positive and negative) affective reaction toward the Church members. Again, we conducted 3 between-group (group membership) and 2 within-group (positive vs. negative affect) ANCOVAs with age, gender, and individual religious orientations as covariates. Age predicted positively positive affect toward the Church members, β = 0.281; p < 0.001. Similarly, individual extrinsic religious orientation predicted positive affect toward the Church members, β = 0.399; p < 0.001. When controlled for the effects, the interaction between group membership and affective reactions toward the Church members was significant (F(2, 190) = 5.852; p = 0.003; η_p² = 0.058; Figure 1).

The “hand only” group showed lower positive affect compared to the “both forms accepted” group (p = 0.009) and marginally higher negative affect compared to the “both forms accepted” group (p = 0.077). Furthermore, the “mouth only” group and the “both forms accepted” group showed more positive than negative affect (p = 0.008, p < 0.001, respectively). The differences between positive and negative emotions were non-significant in the “hand only” group (p = 0.724).

2.3.3. Group Membership and Legitimacy of the Church Authority

Second, we investigated whether group membership (“hand only” vs. “mouth only” vs. “both forms accepted”) resulted in differentiation in the legitimacy of the Church authority. Again, we conducted 3 between-group (group membership) ANCOVAs with age, gender, and individual religious orientations as covariates. Intrinsic religious orientation predicted higher legitimacy of the Church authority (β = 0.269; p < 0.001). The main effect of group membership was significant (F(2, 190) = 5.202; p = 0.006; η_p² = 0.052). The “both forms accepted” group reported higher legitimacy of the Church authority (M = 2.759; SD = 1.073) compared to the “hand only” group (M = 2.273; SD = 1.142; p = 0.019) and the “mouth only” group (M = 2.136; SD = 1.171; p = 0.012).

2.3.4. Intergroup Bias in the Social Perception of Religiosity

Third, we investigated whether group membership (“hand only” vs. “mouth only” vs. “both forms accepted”) predicted the social perception of intrinsic and extrinsic religious orientations to the “hand only” and the “mouth only” groups. We conducted 3 between-group (group membership) and 2 within-group (ascription of religious orientation to the target group, i.e., “hand only” vs. “mouth only” group) and 2 within-group (intrinsic vs. extrinsic religious orientation) ANCOVAs with age, gender, and individual religious orientations as covariates. In accordance with the aim of the study, we were mostly interested in a two-way interaction between group membership and the target group and a three-way interaction between group membership and the target group and the religious orientation. The two-way interaction was significant (F(2, 190) = 9.560; p < 0.001; η_p² = 0.091). The post-hoc Tukey’s test showed that in the “mouth only” group, the effect of significantly lower ascription of religion orientation to the out-group (here, the “hand only” group) was found (M_out-group = 2.211; SD = 0.941 vs. M_in-group = 2.655; SD = 0.918; p < 0.001). Moreover, the level of religious orientation of the “hand only” group estimated by the “mouth only” group was significantly lower than other estimations made by other group members to in-groups and out-groups (p < 0.041). However, this two-way interaction was also qualified by the three-way interaction with the type of religious orientation (F(2, 190) = 11.053; p < 0.001; η_p² = 0.104). The decomposition of this interaction is given in Figure 2.

The effect of out-group favoritism occurred in the “hand only” group. The “mouth only” group (out-group) was described as more extrinsically oriented (M = 2.468; SD = 0.823) compared to the in-group (M = 2.197; SD = 0.621; p = 0.001). In both in-groups and out-groups, the intrinsic religious orientation was estimated as significantly higher than the extrinsic religious orientation (p < 0.001 and p = 0.043, respectively). Intergroup bias was found in the “mouth only” group. The estimated intrinsic religious orientation to the in-group was significantly higher (M = 2.935; SD = 0.965) than intrinsic religious orientation ascribed to the out-group (“hand only”; M = 2.289; SD = 0.910; p < 0.001). The estimated extrinsic religious orientation to the in-group was marginally significantly higher (M = 2.374; SD = 0.792) than extrinsic religious orientation ascribed to the out-group (“the hand only”; M = 2.133; SD = 0.755; p = 0.081). Moreover, intrinsic orientation ascribed to the in-group was higher than the ascribed extrinsic orientation to the in-group (p < 0.001), while the ascribed intrinsic orientation to the out-group did not differ significantly from the extrinsic orientation ascribed to the out-group (p = 0.686).

The pattern of the results suggested that the effect of derogation of the out-group (“hand only”) occurred in the “mouth only” group. However, the effect of the out-group (“mouth only”) favoritism in the domain of extrinsic religious orientation occurred in the “hand only” group. No intergroup bias was found in the “both forms accepted” group. Intrinsic religious orientation was estimated as higher than extrinsic religious orientation both in the “hand only” and the “mouth only” target groups (p < 0.001). The individual extrinsic religious orientation predicted higher ascription of intrinsic and extrinsic orientation to both the “hand only” and the “mouth only” groups (β ranged from 0.170; p = 0.031 to 0.544; p < 0.001). The individual intrinsic religious orientation predicted higher ascription of intrinsic orientation to the “hand only” group (β = 0.218; p = 0.003).

2.4. Discussion

The present study demonstrated that among highly religious Christians in Poland, three separate groups could be distinguished depending on the beliefs about the proper form of receiving Holy Communion. The groups that included persons recognizing both forms of the reception (i.e., on the hand or in the mouth) and persons considering the reception of Holy Communion on the hand as proper were similar in number. A slightly lower number of participants were convinced that Holy Communion should be received in the mouth. Thus, the manner of receiving Holy Communion during the pandemic implied a division among Catholic Church followers in Poland. Moreover, the present study demonstrated that beliefs about the proper form of receiving Holy Communion were related to emotions experienced toward the community of the Catholic Church, affected the legitimacy of the Church authority, and intergroup bias in attribution of religiosity (intrinsic and extrinsic orientation ascription), which emerged in the group recognizing reception of Holy Communion in the mouth as the proper form.

First, those considering reception of Holy Communion on the hand only experienced more negative emotions and fewer positive emotions than the other groups. As a result, their emotional attitude toward the whole religious community was the most ambivalent. This effect could be possibly due to stronger fear of infection experienced in the “hand only” group. Feeling at risk of infection was shown to be predictive for the sanitary restriction (hand washing, social distancing, etc.; Harper et al. 2020). In the present study, we showed that when confronted with behaviors that were less consistent with sanitary restrictions, persons who obeyed these restrictions may feel less positive toward other people and more negative. This effect did not occur among those who did not make the behavior change due to the pandemic (“mouth only”) or those accepting both forms of reception of Holy Communion. Among the last two groups, positive emotions toward the whole Church community were experienced more frequently than negative emotions. This pattern of results may indicate that intragroup relationship conflict could be experienced in the “hand only” group (Jehn 1997).

Second, the present study found differences between the legitimacy of the Church authority among the groups of followers according to their beliefs about the proper form of the reception of Holy Communion. Both the “hand only” and the “mouth only” groups showed lower legitimacy of the Church authority compared to the “both forms accepted” group. Harper et al. (2020) demonstrated that the moral code of authority/respect was positively correlated with behavior change in the pandemic. The present study showed that the legitimacy of the Church authority was higher only when the decision was fully consistent with the followers’ beliefs. Polish episcopal and diocesan authorities suggested that Holy Communion should be received on the hand. However, they did not forbid reception in the mouth. Thus, both the “hand only” and the “mouth only” groups could feel the lack of complete congruence with the decisions of the Church authorities and therefore reported lower legitimacy of their authority.

Third, the beliefs about the proper form of reception of Holy Communion affected some intergroup bias during the pandemic. Two effects were observed in the followers sharing clear beliefs about the reception of Holy Communion. The effect of out-group favoritism occurred among members of the “hand only” group in the case of extrinsic religious orientation ascription to the “mouth only” group. Although this effect seemed to be favoritism, it could also be understood as a subtle derogation of out-groups. Extrinsic religious orientation implies that the person places emphasis on protection, consolation, social status, and group participation (Allport and Ross 1967). Thus, the “hand only” group could perceive the religiosity of the members of the “mouth only” group as based more on the need for consolation and a rigid approach to religious rules, as the one trying to gain social status, as more religious, and as less anxious of infection compared to other followers. On the other hand, this result may be consistent with costly signaling (Sosis 2005). Reception of Holy Communion in the mouth could be understood as a costly signal of commitment to a religious in-group and increased the ascription of religiosity. Future research should investigate in more detail the attributions made by the “hand only” group regarding the “mouth only” group.

The most pronounced intergroup bias occurred in the “mouth only” group and consisted in less favorable estimation (or more uncertain estimation) of the out-group’s religious orientation (both intrinsic and extrinsic). Holy Communion in Poland is generally distributed in one form (wafer on the tongue; Sulkowski and Ignatowski 2020). Therefore, in the “hand only” group, the alteration of this ritual could have been related to the fear of infection and the fear of spreading the virus. Conversely, the lack of modification or alteration of beliefs about the proper form of reception of Holy Communion may have reflected the higher conservatism of the “mouth only” group of followers. The change in beliefs and behaviors connected to the reception of Holy Communion may be regarded by more conservative followers as a sign of weak faith in the healing power of Holy Communion or even as a sign of less reverence for holiness. Thus, those who chose to receive Holy Communion on the hand during the pandemic could be perceived as less religious compared to those who maintained the usual form of reception of Holy Communion. This result was consistent with the theory indicating that alteration of ritual reflects a violation of values represented in the ritual (Stein et al. 2021b). Moreover, Karwowski et al. (2020) demonstrated that elevated social conservatism during the pandemic was predicted by higher anxiety. Crawford (2017) showed that physical threat was stronger among conservatives. We suggest that persons who continued to believe that reception of Holy Communion in the mouth during the pandemic was still proper may experience a specific form of anxiety, namely higher value threat (see Rowatt and Al-Kire 2021). Previous meta-analytic studies showed that higher value threat is related to prejudices (Riek et al. 2006). In the present study, the conservative group did not display more negative emotions toward the whole religious community, but a subtle effect of the ascription of lower religiosity to the out-group relative to the in-group occurred. Thus, intergroup bias in the “mouth only” group may also result from the threat that their beliefs should be modified in order to meet the sanitary measures introduced during the pandemic. Again, the obtained results are consistent with costly signaling (Sosis 2005). People who did not engage in costly signaling were perceived less favorably (e.g., Hall et al. 2015). Thus, their religious orientations were estimated as lower compared to those who maintained costly signaling during the pandemic. Future studies should investigate potential intrapersonal threats experienced by those conservative followers such as fear of disregard for holiness.

The present study showed that individual religious orientation predicted positive affect experienced toward the religious group members, the legitimacy of the Church authority (mainly the intrinsic orientation), and ascription of religiosity to others (mainly the extrinsic orientation). Individual religious orientation did not alter the intergroup effects, which were connected to the social categorization processes related to the different beliefs of followers about the proper form of reception of Holy Communion during the pandemic. However, future studies should investigate personal predictors of beliefs about receiving Holy Communion during the COVID-19 pandemic (e.g., fear of COVID-19; Clark et al. 2020; or moral foundations; Graham and Haidt 2010) and analyze whether these personal variables may moderate the intergroup processes. Previous studies showed that both fear of COVID-19 and moral foundations predicted higher behavior change concerning preventive behaviors (Harper et al. 2020).

3. Study 2

In study 2, we investigated whether the between-religion controversy related to the proper form of receiving Holy Communion during the pandemic could have caused intergroup empathy bias among members of the Polish Catholic Church. We used vignettes describing target persons infected with SARS-COV-2 and represent the same belief (in-groups) or opposite belief (out-group) about receiving Holy Communion during the pandemic. Additionally, we measured the COVID-19 related fear and moral foundations as potential predictors of beliefs about the receiving Holy Communion (Harper et al. 2020). We predicted that the care/harm foundation and fear of COVID-19 would be positively related to perceiving reception on the hand as a more proper form of receiving Holy Communion than in the mouth.

We also investigated individual factors that could foster the examined intergroup bias, namely fear of COVID-19 and moral foundations. The COVID-19 risk perception predicted intergroup bias (e.g., in donation for the fight with the coronavirus pandemic; Li et al. 2020). Moral foundations are innate, modular foundations of moral reasoning with evolutionary roots but also shaped by the social and cultural environment (Graham et al. 2011). The care/harm and fairness/cheating foundations are often mentioned as individualizing foundations, while the loyalty/betrayal, authority/subversion, and sanctity/degradation foundations constitute the category of binding foundations (Graham et al. 2011). Previous studies showed that individualizing foundations predicted positive behaviors and less social exclusion toward out-groups, while binding foundations predicted more negative behaviors and more social exclusion toward out-groups (Hadarics and Kende 2018). Thus, focus on both the COVID-19-related fear and moral foundations may help to better explain the mechanisms of intergroup bias detected in study 1.

3.1. Materials and Methods

3.1.1. Participants and Procedure

One hundred and seventy-nine individuals (143 women and 36 men), all Roman Catholic Christians, participated in the present study. The age of the participants ranged from 17 to 60 years (M = 30.1; SD = 10.6). All of them described themselves as involved in religious practices. One hundred and three participants reported higher education (57.5%), 66 reported secondary education (36.9%), and 5 persons reported lower education (primary or vocational). The sample size was determined as in Study 1 using G*Power (Faul et al. 2007). The required sample size to detect a small effect size (f² = 0.10) in ANOVA with repeated measurements and interactions of between-group and within-group factors (power = 0.80; α = 0.05) was computed as N = 174. Thus, the number of participants in the present study met this criterion. The study was approved by the institutional ethics board (KEUS.98/02.2021).

3.1.2. Measures

The following measures were used:

• Opinions about the proper form of reception of Holy Communion: In order to assess the beliefs about the proper form of reception of Holy Communion, we asked the participants to indicate to what extent a particular form of receiving Holy Communion was proper, safe, and justified during the pandemic. The Likert-type scale used ranged from 0 (Not at all) to 4 (Very much). We also asked the participants to make the same evaluations for other common preventive behaviors: Wearing masks and social distancing (Harper et al. 2020). Reliability of measures of appropriateness of these behaviors (receiving Holy Communion on the hand, in the mouth, wearing masks, and social distancing) was satisfactory, α > 0.850.
• Empathic responding toward a target person: We assessed two types of empathic emotional reactions toward a target person: Empathic concern (responding with compassion and tender feelings toward an observed person) and empathic distress (responding with own distress in response to negative and challenging situations faced by an observed person). In order to measure both emotional reactions, we used adjectives taken from Batson et al. (1987). Empathic concern was measured with the following emotions: Compassionate, softhearted, moved, and warm, while personal distress was measured with the following emotions: Upset, distressed, worried, and troubled. The participants were instructed to report how strongly they felt these emotions toward the target person described in a scenario using a 5-point Likert-type scale ranging from 0 (Not at all) to 4 (Very strongly). Reliability of the empathic concern scale ranged from α = 0.719 to α = 0.817 in various experimental conditions. Reliability of the personal distress scale ranged from α = 0.876 to α = 0.912 in various experimental conditions.
• COVID-19-related fear: The Fear of COVID-19 Scale (Ahorsu et al. 2020) consists of seven items (e.g., “I am most afraid of Corona”; “It makes me uncomfortable to think about Corona”). The participants indicate their level of agreement with the statements using a five-item Likert-type scale ranged from 0 (Strongly disagree) to 4 (Strongly agree). The reliability of the scale was α = 0.799 in the present study.
• Moral foundations: The Moral Foundations Questionnaire (Graham et al. 2011), Polish version: Jarmakowski-Kostrzanowski and Jarmakowska-Kostrzanowska (2016) consists of 30 items and asks the participant to what degree he or she agrees with five moral dimensions: Care/harm, fairness/cheating, authority/subversion, loyalty/betrayal, and sanctity/degradation. There are two sections in the questionnaire: Judgments and relevance. In the first one, the participants rate the importance of each of the criteria when they make moral judgments (e.g., “Whether or not someone did something to betray his or her group”). In the second, the participants rate the degree to which they agree with each of the moral judgments (e.g., “I think it’s morally wrong that rich children inherit a lot of money while poor children inherit nothing”). For each moral dimension, a composite score was formed by taking the average of six items (three items from the first section, three items from the second). Each subscale was reliable in the present study, 0.608 (fairness/cheating foundation) < α < 0.743 (sanctity/degradation foundation).

3.1.3. Procedure

The participants were invited to the study by means of information posted in social media groups gathering people involved in religious movements and teaching religion. They were also invited to share the invitation with other persons. After clicking the link to the study, the participants were informed about the aim of the questionnaire and were asked to accept the terms of participation. First, the participants answer the socio-demographic questions (gender, age, religious identification, and religious beliefs) and questions about the appropriateness of preventive measures used during the pandemics (wearing masks, keeping social distance, receiving Holy Communion). Next, they were asked to indicate which form of reception of Holy Communion was proper (“hand only” vs. “mouth only” vs. “both forms accepted” vs. “spiritual reception of Holy Communion”). Then, we asked the participants to respond to the statement from the Fear of COVID-19 scale and the moral foundations questionnaire. In the next section, experimental manipulation was included. The participants were informed that they would read three vignettes that consisted of statements of persons infected with SARS-COV-2. The participants were informed that these vignettes had been drawn from a database collected in another study. The participants were asked to read each vignette and indicate their feelings toward the target person. The vignettes were constructed very similarly to each other and differed only in terms of the experimental manipulation, which included mentioning that the target person had been receiving Holy Communion during the pandemic on the hand (“on the hand” target) or in the mouth (“in the mouth” target). One vignette did not include any reference to receiving Holy Communion and was treated as the control condition. The participants read and responded to all three vignettes.

3.2. Results

3.2.1. Descriptive Statistics

The majority of participants declared accepting both forms of reception of Holy Communion (N = 72; 40.2%). Fifty-eight accepted receiving Holy Communion on the hand (32.4%), while twenty-seven in the mouth (15.1%) and twenty-two (12.3%) preferred the spiritual form of reception of Holy Communion. Sharing a particular belief about the proper form of reception of Holy Communion was related to the assessment of appropriateness of receiving Holy Communion on the hand or in the mouth (F(3, 175) = 97.827; p < 0.001; η_p² = 0.626). Receiving Holy Communion on the hand was assessed as the most appropriate form by people who shared the belief that during the pandemic, Holy Communion should be received on the hand (M = 3.770; SD = 0.460) compared with other groups of beliefs (p < 0.001 in HSD post-hoc tests). Receiving Holy Communion in the mouth was assessed as the most appropriate form by people who shared the belief that during the pandemic, Holy Communion should be received in the mouth (M = 3.605; SD = 0.599) compared to other groups of beliefs (p < 0.055 in HSD post-hoc tests). People accepting both forms of the reception of Holy Communion assessed both forms as equally appropriate (M_{on the hand} = 2.764; SD = 1.074, and M_{in the mouth} = 2.889; SD = 0.892). People preferring the spiritual reception of Holy Communion assessed receiving it on the hand (M = 2.333; SD = 1.251) as more appropriate than receiving it in the mouth (M = 1.379; SD = 1.071).

Means, standard deviations, and reliability of empathic concern and personal distress reported by the participants in response to experimental vignettes are presented in

Table 2. Descriptive statistics for the study 2 variables.

Variable	1	2	3	4	5	6
Empathic concern toward “on the hand” (target person)
Personal distress toward “on the hand” (target person)	0.492
Empathic concern toward “in the mouth” (target person)	0.778	0.368
Personal distress toward “in the mouth” (target person)	0.438	0.841	0.477
Empathic concern toward “control” (target person)	0.867	0.510	0.835	0.506
Personal distress toward “control” (target person)	0.451	0.942	0.375	0.869	0.526
M	2.242	1.975	2.028	1.687	2.161	1.844
SD	0.880	1.116	0.992	1.155	0.984	2.000
α	0.719	0.876	0.817	0.911	0.816	0.912

Note. All correlation coefficients are significant at p < 0.001.

.

3.2.2. Empathy Bias

Next, we investigated whether group membership (“hand only” vs. “mouth only” vs. “both forms accepted” vs. “the spiritual reception”) predicted the empathic concern or personal distress toward the target person infected with SARS-COV-2 in three experimental conditions (“on the hand” vs. “in the mouth” vs. control condition). We conducted four between-group (group membership) and three within-group (the target person in a vignette) and two within-group (empathic responding variable: Empathic concern vs. personal distress) ANCOVAs with age and gender as covariates.

Table 3. The effects of belief group membership and target person predicting two forms of empathic responses (empathic concern and personal distress) controlled for covariates.

Effect	Num DF	Den DF	F	P
Gender	1	167	6.178	0.010
Age	1	167	3.879	0.051
Beliefs about the proper form of reception of Holy Communion (BPHC)	3	167	5.662	0.001
Target person in a vignette (TPV)	2	334	2.543	0.080
TPV × Gender	2	334	0.012	0.883
TPV × Age	2	334	0.969	0.380
TPV × BPHC	6	334	3.217	0.004
Empathic response (ER: EC vs. PD)	1	167	4.291	0.040
ER × Gender	1	167	3.054	0.082
ER × Age	1	167	0.014	0.907
ER × BPHC	3	167	1.410	0.241
TPV × ER	2	334	2.396	0.093
TPV x ER × Gender	2	334	2.371	0.095
TPV × ER × Age	2	334	2.037	0.013
TPV × ER × BPHC	6	334	2.367	0.030

summarizes the effects of membership to groups of different belief about the proper form of the reception Holy Communion during the pandemic predicting how compassionate or personally distressed participants reacted toward target persons in vignettes. All of the main effects and lower-order interactions were qualified by the belief about the proper form of reception of Holy Communion × target person × empathic response type 3-way interaction (F(6, 334) = 2.367; p = 0.030; η_p² = 0.041).

In order to unpack the 3-way interaction (Figure 3), we compared in-group to out-group slopes across the different combinations of factors using HSD post-hoc tests.

Intergroup empathy bias emerged among people who believed that receiving Holy Communion on the hand was the proper form. They reported higher empathic concern toward an in-group target person (M = 2.325; SD = 0.802) compared to an out-group target person (“in the mouth”; M = 2.026; SD = 0.975; p < 0.001 in HSD post-hoc test), and higher personal distress toward an in-group target person (M = 2.189; SD = 1.094) compared to an out-group target person (“in the mouth”; M = 1.930; SD = 1.130; p < 0.001 in HSD post-hoc test). Intergroup empathy bias did not emerge among people preferring receiving Holy Communion in the mouth. People accepting both forms of reception of Holy Communion reported higher personal distress toward a target person described as receiving Holy Communion on the hand (M = 1.972; SD = 1.064) compared to a target person described as receiving Holy Communion in the mouth (M = 1.688; SD = 1.132; p < 0.001 in HSD post-hoc test). People preferring the spiritual reception of Holy Communion reported lower empathic concern (M = 1.821; SD = 0.807) and personal distress (M = 1.607; SD = 1.051) toward a particular out-group target, namely a person described as receiving Holy Communion in the mouth during the pandemic, compared to the second out-group (a target person who received Holy Communion on the hand; M_{empathic concern} = 2.202; SD = 0.921; and M_{personal distress} = 2.048; SD = 1.030, respectively) and control target (M_{empathic concern} = 2.381; SD = 0.875; and M_{personal distress} = 2.012; SD = 1.244.

3.2.3. Moral Foundations and Fear of COVID-19 as Predictors of Opinions about the Proper Form of Receiving Holy Communion

In order to examine the hypotheses posited in the discussion of study 1 about the predictors of opinions concerning the forms of receiving Holy Communion during the pandemic, we conducted additional regression analyses. We regressed opinions about the appropriateness of receiving Holy Communion on the hand and in the mouth onto moral foundations and COVID-19-related fear. We also regressed two alternative preventive measures during the pandemic (wearing masks and keeping social distance) in order to compare preventive measures that are related or unrelated to religious content (

Table 4. Descriptive statistics for the study 1 variables.

Predictor	Wearing Masks		Social Distancing		Holy Communion on the Hand		Holy Communion in the Mouth
	β	sr2	β	sr2	β	sr2	β	sr2
Gender	−0.050	0.002	−0.092	0.007	0.086	0.006	0.110	0.010
Age	0.067	0.004	0.096	0.008	0.046	0.002	0.105	0.010
Fear	0.287 ***	0.076	0.316 ***	0.092	0.155 *	0.022	−0.345 ***	0.110
Care/harm	0.367 ***	0.077	0.264 ***	0.040	0.146	0.012	−0.257 **	0.038
Fairness/cheating	−0.095	0.005	−0.091	0.005	0.045	0.001	−0.123	0.009
Loyalty/betrayal	−0.194	0.018	−0.088	0.004	−0.146	0.010	0.110	0.006
Authority/subversion	0.198	0.016	0.095	0.004	−0.135	0.007	0.151	0.009
Sanctity/degradation	−0.187	0.019	−0.190	0.019	−0.104	0.006	0.207 *	0.023

Note. RO—religious orientation; PA—positive affect; NA—negative affect; LCA—legitimacy of Church authority; IRO—intrinsic religious orientation; ERO—extrinsic religious orientation. sr²—semi-partial correlation. * p < 0.05; ** p < 0.01; *** p < 0.001.

).

The regression model predicted 16.34% of variance in mask wearing (F(8, 164) = 5.208; p < 0.001), and significant positive predictors were COVID-19 related fear and care/harm foundation. The regression model predicted 12.85% of variance in social distancing (F(8, 164) = 4.170; p < 0.001), and significant positive predictors were again COVID-19-related fear and the care/harm moral foundation. The regression model predicted 10.55% of variance in assessment of appropriateness of reception of Holy Communion on the hand (F(8, 164) = 3.536; p < 0.001), and a significant positive predictor was COVID-19-related fear. The regression model predicted 29.95% of variance in the assessment of appropriateness of reception of Holy Communion in the mouth (F(8, 164) = 10.191; p < 0.001), and significant negative predictors were COVID-19-related fear and the care/harm foundation, while the sanctity/degradation foundation was a positive predictor.

People sharing different beliefs about the proper form of receiving Holy Communion (“hand only” vs. “mouth only” vs. “both forms accepted” vs. “the spiritual reception”) also differed in the level of COVID-19-related fear (F(3, 175) = 3.381; p = 0.022; η_p² = 0.055) and moral foundations (Wilks’ lambda = 0.775; F = 3.051; p < 0.001). The “mouth only” group reported a lower level of fear (M = 0.545; SD = 0.454) compared to the “hand only” group (M = 1.012; SD = 0.756; p = 0.036 in HSD post-hoc test). The “mouth only” group reported lower care/harm foundation (M = 4.784; SD = 0.938) compared to the “spiritual reception” group (M = 5.303; SD = 0.681; p = 0.040 in post-hoc HSD test). The “spiritual reception” group reported a lower authority/subversion foundation (M = 3.424; SD = 0.960) compared to the “mouth only” group (M = 4.173; SD = 0.944; p = 0.014 in post-hoc HSD test) and “both forms accepted” (M = 4.102; SD = 0.732; p = 0.033 in post-hoc HSD test).

3.3. Discussion

Intergroup empathy bias is the tendency to empathize less with out-group relative to in-group members (Cikara et al. 2014). Previous studies demonstrated that intergroup empathy bias was documented among real, social groups, such as racial or political groups (Cikara et al. 2011). The present study demonstrated that people who believe that Holy Communion should be received on the hand during the pandemic reported higher empathy toward the in-group relative to the out-group (“mouth only”). People who believe that Holy Communion should be received spiritually during the pandemic reported dampened empathy toward a particular out-group (“mouth only”). Thus, similarly to study 1, intergroup bias appeared as a consequence of differentiation in beliefs about the proper form of receiving Holy Communion. However, intergroup bias emerged only in particular subgroups.

Fear of COVID-19 and the care/harm foundation were positive predictors of the approval of preventive behaviors (mask wearing, social distancing), also in the religious domain (receiving Holy Communion on the hand), during the pandemic. These findings were consistent with previously demonstrated positive associations between both fear of COVID-19 and the care/harm moral foundation and preventive behaviors during the pandemic (Harper et al. 2020; Qian and Yahara 2020). The approval of reception of Holy Communion in the mouth during the pandemic was positively predicted by the sanctity/degradation foundation, while negatively predicted by the care/harm foundation and COVID-19-related fear. However, people sharing different beliefs about the proper form of receiving Holy Communion did not differ in the level of endorsement of the sanctity/degradation moral foundation. Thus, a person’s decision about the form of reception of Holy Communion seemed to be a consequence of two processes: (a) A moral conflict between the care/harm moral foundation and the sanctity/degradation foundation; and (b) the level of COVID-19-related fear. People accepting reception of Holy Communion in the mouth reported a relatively low level of fear of COVID-19 and a relatively lower care/harm foundation, thus their beliefs could have been more associated with the sanctity/degradation foundation. A higher level of fear of COVID-19 combined with a higher importance of the care/harm foundation could have been the reasons for persons who preferred receiving Holy Communion on the hand only.

Less is known about the two remaining groups of beliefs about receiving Holy Communion. People accepting both forms could have been avoiding a decision about the proper form of receiving Holy Communion. Among people who prefer the spiritual form of the reception Holy Communion, the care/harm foundation could have had the most pronounced impact on the decision to avoid risky behavior (receiving Holy Communion on the hand). Members of the “spiritual reception” group reported the lowest empathy for the target person who was infected and was described as receiving Holy Communion in the mouth during the pandemic. This may indicate that members of this group could have seen receiving Holy Communion in the mouth as violation of the care/harm foundation in order to prevent the sanctity/degradation foundation. This could have caused less empathy toward the target person who was infected and had tended to receive Holy Communion in the mouth before.

4. General Discussion

The present study demonstrated that the belief about the proper form of reception of Holy Communion during the COVID-19 pandemic could be regarded as a source of a subtle intragroup relationship conflict in the Polish Catholic Church (Jehn 1997). This division is manifested in more negative and less positive experiences toward the Church members in persons who believe that Holy Communion should be received on the hand. The second manifestation was lower legitimacy of Church authority among those who believed that the proper form of reception of Holy Communion was on the hand only or in the mouth only. This result showed that decisions of the Church authorities may meet with resistance when they are inconsistent with personal beliefs of the followers. Third, subtle effects of the out-group derogation were observed in study 1, in the “hand only” and the “mouth only” groups. These intragroup biases consisted of (a) the ascription of higher extrinsic religiosity to the “mouth only” group by persons recognizing reception of Holy Communion on the hand as the only proper for and (b) the ascription of lower religiosity (both lower intrinsic and extrinsic orientation) to the “hand only” group by persons recognizing reception of Holy Communion in the mouth as the proper form. The obtained results were also consistent with costly signaling (Sosis 2005). The “in the mouth” group was perceived as more religious by those who decided not to engage in costly signaling (reception of Holy Communion in the mouth) during the pandemic. Intergroup empathy bias was observed in study 2 in reporting less empathy toward the infected out-group (among the “hand only” group toward the “mouth only group”). Similarly, intergroup bias was also observed among the “spiritual reception” group toward the out-group who potentially violated the care/harm moral foundation by religious behavior (“mouth only” group). The obtained result seems to be inconsistent with costly signaling, which has demonstrated that people engaging in religious costly signals are trusted more (Hall et al. 2015). The intragroup relationship conflict demonstrated among Roman Catholics in Poland during the pandemic may be regarded as caused by the changed status of reception of Holy Communion in the mouth as a costly signal. The cost of reception of Holy Communion in the mouth during the pandemic also involved the threat of infection or of infecting others. Thus, the costly signal became more complex. For example, people receiving Holy Communion on the tongue expressed both praising the value of the Body of Christ but also lack of compassion for those who, due to fear of infection, preferred receiving Holy Communion on the hand. Our results partially confirmed the theory of ritual alteration as moral violation (Stein et al. 2021a, 2021b). The subtle derogation of the out-group indicated that negativity was expressed toward those who altered the ritual but also toward those who did not alter it.

The current study demonstrated that different reactions toward ritual alteration due to preventive measures used during the pandemic can illustrate a process of ritual (re)negotiation (Hüsken and Neubert 2012). Since ritual failure is always a failure from someone’s perspective (Grimes 1990), the intergroup negativity shown in two studies may indicate that Catholics who recognized receiving Holy Communion in the mouth only can perceive other followers as violating the ritual, which causes their subtle negativity toward them (i.e., perceiving them as less religious). On the other hand, resistance of more traditionally oriented Catholics can be perceived as unjustified in the light of the pandemic circumstances (e.g., danger of the virus spreading) and as an act of disobedience toward the Church authorities by the “hand only” group. The process of renegotiation of the ritual of receiving Holy Communion during the pandemic may, in turn, lead to a ritual change in the Catholic Church in Poland.

The obtained results demonstrated that even Holy Communion, whose name is partially derived from the word meaning community, could divide religious followers as a result of basic social identity processes (Tajfel and Turner 1986). Thus, Church authorities, clergy, and Church members should make additional efforts to focus on the unity of the Catholic Church during the pandemic in order to avoid consequences of intergroup bias.

The present studies have some limitations. One of the limits of this study is its decision not to address any underlying issues of sacramental theology, which are important to understand decisions regarding the form of receiving Holy Communion but remain beyond the scope of this work. Second, although the official position of the Polish Episcopate indicated that receiving Holy Communion on the hand is an appropriate form during the pandemic, the positions of particular Church authorities differed between dioceses. The Archbishop of Szczecin asked priests and the faithful not to promote receiving Communion on the hand (https://kuria.pl/instytucje/Slowo-pasterskie-Ksiedza-Arcybiskupa-Metropolity-na-III-Niedziele-Wielkiego-Postu-15-marca-2020-r_4008, accessed on 27 February 2021). This can suggest that the pattern of intergroup perception between the followers who changed the form of receiving Holy Communion and those who did not alter their form of reception of Holy Communion can be affected by the suggestions of local Church authority. Thus, future studies should address particular cases such as the Diocese of Szczecin. The studies were conducted among highly religious persons, which may limit the generalization of the results. However, reception of Holy Communion seems to be related to the frequency of participation in religious services, thus we investigated the group that was potentially frequently exposed to the discussion about the proper form of the reception of Holy Communion and to the situations of facing different forms of reception during the pandemic. Moreover, the participants were more religiously involved, which may promote intergroup bias, which is stronger when the group identification is more pronounced (Demoulin et al. 2009). Second, the study design did not include the ascription of religiosity to those who accepted both forms of reception of Holy Communion or empathy experienced toward that group. However, we excluded this reference group from the study design due to the fact that reception of Holy Communion is a dichotomous behavior (a follower can receive Holy Communion only on the hand or in the mouth). However, future studies should investigate how consistency in the followers’ behavior regarding reception of Holy Communion during the pandemic may affect the ascription of religiosity. It can be predicted that those with an inconsistent approach may be perceived less favorably because their behavior may be considered as a violation of the values of a particular group. Lastly, the differences between people who shared different beliefs about the proper form of receiving Holy Communion may have also resulted from previous differences in beliefs or orthodoxy. Thus, the intragroup processes examined in the present studies could have become apparent in the context of the pandemic, but not due to the pandemic.